Department of Sociology

Umeå University

Organisational characteristics and psychosocial working conditions in different forms of ownership

by

Chatrine Höckertin

Doctoral theses at the Department of Sociology Umeå University

No 49 2007

Department of Sociology Umeå University Thesis 2007 Printed by Print & Media April 2007 © Chatrine Höckertin ISSN 1104-2508 ISBN 978-91-7264-309-3

••

Höckertin, Chatrine. Organisational characteristics and psychosocial working conditions in different forms of ownership. Doctoral Dissertation in Sociology at the Faculty of Social Sciences, Umeå University, 2007. ISBN 978-91-7264-309-3 ISSN 1104-2508

ABSTRACT The main aim of this thesis has been to compare psychosocial working conditions in workplaces with different forms of ownership, i.e. public, private and cooperative. A second aim has been to study how organisational characteristics of relevance for psychosocial working conditions (in terms of management control strategies and prerequisites for management) are manifested in these ownership forms. The empirical data is based on structured interviews with managers at 60 workplaces within the service sector and on a questionnaire to all employees working in the participating workplaces, resulting in a set of 1384 individuals. An additional seven interviews with first-line managers within geriatric care were also conducted for the last study. The results show that employees in cooperatives perceived that they had better opportunities to influence decisions concerning the workplace as a whole, although there were also results showing advantages for public and private employees. Regarding opportunities for employees to influence their own work situation, there were no differences between the ownership forms. Differences were found in the prerequisites for first-line geriatric care managers. As a result of an earlier organisational change, the public managers were now further away from the strategic level and had to focus on daily, operative work tasks, while simultaneously also being responsible for keeping within the budget. The private managers, on the other hand, having group leaders to deal with the daily work concerning personnel and operations, could focus more on strategic work related to financial results in terms of planning and follow-up of the budget. One conclusion is that there are certain differences in both psychosocial working conditions and organisational characteristics between the ownership forms, but when the comparisons were restricted to only one type of service, in this case the provision of care, it is rather the similarities within the care organisations, regardless of ownership form, that are most pronounced. Key words: Psychosocial working conditions, prerequisites for management, management control strategies, cooperative, public sector, ownership, welfare services, geriatric care.

••

Preface Having finally reached the end of this shift, I feel that I would like to take the whole world in my arms for the biggest hug ever made! This is of course not practically possible, but there are still a lot of people who have supported me in different ways in my work and my life, who deserve special mention! When I started work at the National Institute for Working Life (NIWL), when it was established in Östersund in 1998, Katarina Grut and Stig Vinberg gave me the initial support that facilitated for my employment. Having the opportunity to take part in the establishment of a new public institution is an interesting but demanding process. This dynamic period has been both stimulating and fun – largely because of such great colleagues. Thank you, all my good friends and workmates at the NIWL in Östersund and Stockholm! I would also like to express my gratitude to Ewa Wigaeus Tornqvist and Mikael Nordenmark for being such supportive managers during my final year as a doctoral student. Special thanks to my workmates Lisa Sundin, Anette Forsberg and Ylva Fjell, for sharing the same topics of joy and anxiety, and to Anette and Lisa for being such good friends and colleagues since the very beginning! Parallel to putting the finishing touches to this thesis, I and my colleagues now sadly have to disband, pack and move due to the closure of the NIWL. May all the knowledge in working life research that has been built up during several decades live on in new constellations and keep contributing to good working conditions! I am very grateful to my supervisors for reading my manuscripts, and for their wise advice, support and inspiration. It has been a pleasure working with you – my sincere thanks to you all: Jonas Höög at Umeå University, Roine Johansson at Mid-Sweden University and Annika Härenstam at the NIWL in Stockholm. In my second year as a PhD student I had the opportunity to take part in a research project where Annika Härenstam was involved. She became my inspiring supervisor in the daily, operative work, and has supplied me with great knowledge and the best possible balance of guidance and freedom! Project leader Staffan Marklund, with his knowledge and enthusiasm, has created the right “everything-is-possible” atmosphere, and Staffan, Annika and my doctoral colleague and friend Malin Bolin, have contributed to shaping a great research climate that helped me forward - thank you for bringing me in! Another important friend and colleague in this connection is Maria Nordin – my warmest thanks to you and Malin for all the scientifically and socially rewarding sessions in Börjesjö, Duved and Stockholm! Other valuable members in the project have been Peter Westerholm, Monika Söderholm, Tomas Lindh, Maud Hagman, Anders Knutsson, Lars Alfredsson and the whole WOLFF team. All those employees and employers at the participating workplaces constituting the data for this thesis also deserve my sincere gratitude – without your participation it would have been much less stimulating writing the thesis! I would like to thank Annika Härenstam, Tor Larsson, Staffan Marklund, Tina Forsberg Kankkunen and John Ylander for their valuable contribution as members of the project “Power over working conditions”, which constituted the basis of my fourth study. The “Orgdok” group at NIWL also deserves a big thank you for organising seminars, providing the opportunity to receive constructive feedback on drafts and manuscripts, and for creating a good, academic environment. It has been worth all those journeys to Stockholm to participate in the sessions!

I would also like to thank Lars-Erik Wolvén, Mid-Sweden University, for support and advice during the initial year of my doctoral work. Thanks also to librarians at the National Institute for Working Life and Mid-Sweden University. Thank you, Hilary Hocking, for your fast and professional work with language revision! Special thanks to Eva Bejerot at the NIWL for critical reading of my manuscript prior to my final seminar, which helped improve my work. Thanks also to Mikael Nordenmark, Stig Vinberg and Lena Karlqvist at the NIWL, and to Björn Halleröd, Stefan Svallfors and Olle Persson at the Department of Sociology, Umeå University, for constructive comments and advice on the final version of the manuscript, and also to Maritha Lundgren, who has supplied invaluable help in the final work of editing the manuscript. I spent several childhood and teenage summers with my relatives Alva and Bertil Daneholt in their summer house on the west coast of Sweden. I remember my early fascination for Bertil’s research work, although the study of molecular genetics seemed rather vague to me. However, Bertil’s enthusiasm influenced me and I remember being struck by his true work dedication and involvement! Thank you, Bertil, for that little seed that inspired me to go for research! Working and living in Östersund, belonging to a project that is based in Stockholm and with my department located in Umeå, has demanded a lot of travelling and planning, especially in terms of combining working life with a family life! This would not have been possible without the invaluable time, help and shared commitment from my family and relatives. You really deserve a preface of your own, so the rest of the page is for you! My grandmother, Irma, was the one who showed me how to read and taught me to seek knowledge as well as the importance of always trying to do my best. At times when I haven’t achieved the best results, you have always been eager to show me unconditional love and affection, so this thesis is for you! It’s wonderful to hear you say “just call me if you need any help in the afternoons when the girls come home from school” – such a helping hand is invaluable, and for me it is especially amazing since you have just turned 83! Thank you for always being there for me! My parents, Ylva and Hans have given me a secure and stable childhood, and always, always been there for me – when I was younger and still today! Thank you for encouraging me in my work but also for all your practical help with everything from picking up grandchildren from their athletics classes to helping us build our winter cottage in the Åre mountains. You know that without your invaluable help there would have been no thesis! Mats, Liv and Lise – now that I’ve finally reached my goal this time round, we can start preparing to enjoy a smooth landing! Let’s buy that dog, go for that long vacation or why not just take a short break, make the most of the early spring sunshine and seize the day! I owe you so much in terms of time and involvement, and I look forward to being able to take a more active part in the activities of the hectic weekdays! Thank you all so much for being there with me, and for having truly shared all the joy with me – it’s really something to have two daughters who go crazy over an accepted manuscript! It’s been wonderful having the three of you as my nearest and dearest during this time! Chatrine Höckertin Östersund, a sunny day in March 2007

••

To Irma, my grandmother

••

Contents Introduction 1

Aim and questions 2 Main perspectives 3

Characteristics in different forms of ownership 4

New ownership forms in welfare services in transiton 7

Changes in organisational characteristics 8

Psychosocial working conditions 10

Summary 12 Design, method and description of data 13 Participants and procedure 13 Design of the thesis and the included studies 15 Empirical findings 21 Psychosocial working conditions (study I and III) 21 Management control strategies (study II) 23 Prerequisites for management (study IV) 24 Discussion 27 References 33

The Attached Articles Article I Höckertin, C. & Härenstam, A. (2006) The Impact of

Ownership on Psychosocial Working Conditions: A Multilevel Analysis of 60 Workplaces. Economic and Industrial Democracy, vol 27(2), 245-284.

Article II Höckertin, C., Johansson, R. & Härenstam, A. (2006)

Management control strategies in public, cooperative and private services. Submitted.

Article III Höckertin, C. (2007) Control and participation for

employees within geriatric care - does ownership make a difference? Accepted for publication in WORK: A Journal of Prevention, Assessment & Rehabilitation.

Article IV Höckertin, C. (2007) Organisational prerequisites and

discretion for first-line managers in public, private and cooperative geriatric care. Submitted.

••

1

Introduction The main interest in this thesis concerns potential differences in psycho-social working conditions for employees in different forms of ownership. The research interest was born in the practical reality of a parent coopera-tive which I had the opportunity to be involved in, and also, as chairman, had the main responsibility for during one year. Compared with the pub-lic childcare our family had just left, I was fascinated by the way in which decisions could be taken easily and at short notice, and how the small-scale business concept created a high degree of transparency, and made employees and members feel that they made a difference and were able to influence activities and decisions in the workplace. It was also an atmos-phere characterised by collective efforts, where most people worked for the common good. At the same time I also noticed some difficulties asso-ciated with the cooperative ideals of fully democratic behaviour, and that the small size and limited resources made it more difficult to deal with work environment issues systematically.

One year after my chairmanship in the cooperative, a new work op-portunity related to cooperatives appeared in a research project at the National Institute for Working Life, where the aim was to map the preva-lence of new cooperatives in Sweden and the potential for new jobs within these cooperatives. While I was working in the project, I came in contact with several persons – researchers and practitioners, professionals and non-professionals – who claimed that cooperatives were better work-places than the public alternatives due to their flat structure and non-hierarchical methods. These arguments inspired me to go deeper into these issues in a systematic way. Can a particular ownership form create better working conditions per se? Or is it more a result of a strong ideo-logical belief that supports these opinions? Having also previously worked both in the public sector and in private firms, the idea of compar-ing the three ownership forms and exploring the different prerequisites for working conditions inspired me. In doing this, I had to take type of service into account, since production in e.g. the public sector, is to a great extent related to welfare. For economic associations, which are the most common juridical form for cooperatives, offshoots from the public sector are mentioned as having the best potential for growth (Statskon-toret, 1997).

Organisational characteristics, such as how the organisation has de-signed its management control strategies, e.g. in terms of formalisation,

2

placement of responsibility and the use of performance measurements, have an indirect effect on psychosocial working conditions. Another im-portant issue in the organisational context is the prerequisites for man-agement, e.g. the relationship between available resources and demands, and these aspects affect the manager’s discretion which is also important for the working conditions of all employees. Therefore, these two dimen-sions are also included in two studies and in relation to different forms of ownership.

Aim and questions The overall aim of the work constituting this thesis is to compare psycho-social working conditions in organisations with different forms of owner-ship and in different services. A second aim is to explore how the organi-sation’s management control strategies and prerequisites for management are manifested in relation to different forms of ownership. The following questions are raised in the separate studies.

- Are there any differences in perceived psychosocial working condi-

tions for employees working in service organisations in different forms of ownership (public, private, cooperative and public enterprises)? Does type of service make any difference in this respect?

- Are there any differences in how public, private and cooperative or-

ganisations design their management control strategies? Are there any differences between services in general and care organisations?

- Are there any differences in how employees in geriatric care in dif-

ferent forms of ownership perceive their control and participation? - What are the prerequisites for first-line managers within public, pri-

vate and cooperative geriatric care?

This thesis has the following disposition. In the following chapter there is a description of the main perspectives and themes relevant for and in-cluded in the separate studies constituting the thesis. The next section deals with methods, questions related to the sample and the design of the included studies. This is followed by a section where the main empirical findings are presented, and finally there is a chapter discussing these find-ings. The four studies on which the thesis is based are attached at the end.

••

3

Main perspectives In this chapter there is a description of the theoretical perspectives on which the thesis is based. This includes certain main features of different forms of ownership and the development of welfare services during the last few decades, organisational characteristics (in terms of management control strategies and prerequisites for management) and psychosocial working conditions. The description is rather short but nevertheless pro-vides a summary of the main perspectives. The included articles present a more detailed theoretical frame in relation to each research question. Figure 1. Theoretical model of the central variables and themes that constitute the thesis.

DETERMINANTS ORGANISATIONAL PSYCHOSOCIAL CHARACTERISTICS WORKING CONDITIONS

Different forms of ownership, i.e. public, cooperative and private, consti-tute the independent variable that is central in all four studies. Due to differences in goals, motives and certain prerequisites, this might create differences in psychosocial working conditions, here studied in terms of demands, control, participation, psychosocial climate and manager’s dis-cretion. Since the type of service, e.g. caring, is not equally distributed between ownership forms, this is also an important variable to take into account.

Ownership Type of service (focus on wel-

fare)

Management control strategies

Prerequisites for management

Manager’s discretion

Demands Control

Participation Psychosocial climate

4

One aspect that affects psychosocial working conditions indirectly is the way in which the organisation has designed its management control strategies (illustrated by a dashed arrow in the model). The management can choose to organise the work in different ways, and with different forms of control functions more or less integrated, e.g. by the formal structure in terms of centralisation and formalisation. In a changing work-ing life with an increasing service sector and new management ideals, more modern strategies for control and supervision have become preva-lent. Measurements of performance, decentralisation of responsibility and an increased emphasis on the customer as the main actor, who then be-comes an active part in the production process, are examples of this de-velopment. Depending on how these management strategies are designed, the impact on psychosocial working conditions can vary. Another impor-tant aspect of psychosocial working conditions concerns the prerequisites for management in an organisation. The combination of responsibility, demands, resources and being able to communicate with superiors are some of the aspects that shape the manager’s discretion, which is of indi-rect importance for working conditions (illustrated by another dashed arrow in the model). In the following, each of the boxes/themes in the model will be briefly described.

Characteristics in different forms of ownership Different forms of ownership (i.e. forms of enterprise) have certain char-acteristics that create heterogeneity and distinctions between them. In the following section there is a schematised table, which illustrates how cer-tain characteristics are expressed in public, cooperative and private or-ganisations respectively. The categorisation constitutes simplified ideal types, where differences between ownership forms have been taken to an extreme in order to elucidate these dissimilarities and to illustrate some specialities within each ownership form. In practice, organisations can rather be seen as hybrids, where certain characteristics can be located on a continuum (Rothschild-Whitt, 1979).

The description is based on the work of Nilsson (1986) and Sjöstrand (1993), where systematisations have been made in order to compare dif-ferent forms of ownership in several aspects. Cooperative characteristics can vary, depending on whether it is a consumer-, producer- or worker cooperative, and the new cooperation that has arisen since the middle of

••

5

the 1980s also differs from the more established cooperation in several aspects, e.g. with more pronounced activity from the members (Küchen, 1994). However, in the summary below the different kinds of coopera-tives are treated as one overall category. Also in private firms there are large differences in these characteristics in practice, e.g. depending on the size of the company. For the workplaces constituting the empirical data used here, the description can nevertheless be useful on a general level. Table 1. Organisational characteristics in public, cooperative and private owner-ship forms.

Ownership form/ Dimensions

Public Cooperative Private

Principal goals Political

Social Economic

Organisational form Hierarchical organisation

Non-hierarchical organisation

Low level of hierarchy

Degree of formali-sation

Strong formalised work process

Low formalised work process

Non-formalised work process

Selection of man-ager made by

Local politician (or delegated)

Members Owners

Type of communication

Primarily vertical Complex, primarily horizontal

Simple but encompassing

Public organisations exist in order to fulfil goals based on political deci-sions. The political management is of central interest and the goals of the organisations, e.g. an elderly care institution, can be seen as sub-goals in the greater political agenda. The typical public organisation is hierarchi-cal with a strongly formalised work process. Local politicians appoint the top managers, who in turn are responsible for recruiting lower level man-agers. Communication is often vertical, from the top down with demands, and from the bottom up and subordinated.

In the cooperatives, the principal goals are based on social foundations and are expressed in the way the cooperatives are often started – a group of individuals discovers a certain problem or need, which they believe they can fulfil by collective efforts. There is a non-hierarchical organisa-tion and a low degree of formalisation of the work process. Managers are selected and chosen by the members, and the type of communication is characterised by complexity and is primarily horizontal. Cooperatives are sometimes said to be something in between the public and private forms, set to fulfil both social and economic goals simultaneously (Pestoff,

6

2000), but according to Nilsson (1986) cooperative organisations are rather characterised by influences from both private and public simulta-neously and as a result of these similarities, cooperatives also have cer-tain characteristics that distinguish them from the other two. Cooperatives are often referred to as belonging to the third sector (Miettinen, 2000). On the international official homepage for cooperatives (www.ica.coop), which is an established and widespread arena for cooperative activity, the specific values related to cooperatives are formulated: “Cooperatives are based on the values of self-help, self-responsibility, democracy, equality, equity and solidarity. In the tradition of their founders, co-operative members believe in the ethical values of honesty, openness, social re-sponsibility and caring for others”. The principles for influence follow the cooperative tradition of “one member, one vote”.

The goals of the private firms are primarily economic and based on fi-nancial resources coming from the owners. They can choose the kind of activity they prefer to be involved in, since they do not have any public obligations. There is a low level of hierarchy and the work is non-formalised. The manager in the private firm can be any citizen or an owner of the firm. In those organisations, communication is often simple but all-embracing, and degree of influence is directly related to the amount of stocks and shares held by the owner.

Surprisingly little research has been done comparing public, coopera-tive and private ownership forms in the same analyses – in any respect. In a recent literature search for studies on comparisons between the three ownership forms (regardless of what they had been compared against) there were almost no studies at all. The search was conducted in July 2006 in ten of the largest international databases1. With the three search-terms public, cooperative and private, only one hit was found where a comparison was made, but it was not relevant in relation to working con-ditions of any kind.

1Sociological Abstract and Social Services Abstract, Business Source Premier, Psych

Info, PAIS, Econlit, ASSIA: Applied Social Sciences Index and Abstracts, Academic Search Elite, Science Direct, Social Sciences Citation Index and Pub Med.

••

7

New ownership forms in welfare services in transition Welfare services are one part of the public sector and an area that has been targeted for major restructuring and changes during the last few decades. The changes, which have taken place in most European coun-tries, first started as an ideological movement in order to improve what was seen as an ineffective public sector. Many of the changes that have taken place can be gathered under the wide term of New Public Manage-ment, NPM, and can be described as a modernisation strategy aiming at increasing efficiency and productivity by imitating ideals concerning competition and management from the private sector (Forssell & Jansson, 2000; Hood, 1991; Hood, 1995). This development, sometimes referred to as a marketisation of the public sector, also gained ground in Sweden, where it has been driven more offensively than in other Scandinavian countries but less so compared with other European countries, e.g. Great Britain (Montin, 1997; Trygegård, 2001). After the political shift to a non-socialist government in the Swedish election in 1991, this develop-ment was intensified. Changes in the legislation in 1992 made it easier for private actors to enter the public welfare sector. Geriatric care is a good example to illustrate this development, which has resulted in a decentrali-sation of responsibility from the state down to the municipalities, thereby increasing their discretion. In addition, there has also been an increased economic burden, since there was a movement of responsibility for long-term care and service for those whose medical treatment was completed, from the county councils to the municipalities (Borell & Johansson, 1998), resulting in 31,000 beds being transferred from the county council to the municipalities (Runesson & Eliasson-Lappalainen, 2000).

The changes in legislation that facilitated for increased privatisation have changed the map of welfare service producers in Sweden, with an increase in both private and cooperative actors. Between 1993 and 2000 the number of employees within private care for the elderly and function-ally disabled increased by more than 400% (Trydegård, 2001). This de-velopment has many parallels to welfare pluralism (Dahlberg, 2005; Johnson, 1987; Lund, 1993), a concept focusing on the idea that welfare services can be equally provided by a combination of four separate sources: the statutory, the commercial, the voluntary and the informal. In an empirical practice the statutory can be translated to the local govern-ment and municipality; the commercial refers to private for-profit actors; the voluntary relates to non-paid, non-profit work, but also includes paid

8

work conducted in the voluntary sector. Criticism has been raised con-cerning the use of the term voluntary, because of the associations with only unpaid work (Dahlberg, 2005; Wijkström & Lundström, 2002). A related term, often used in Sweden is the third sector, referring to some-thing different from the public and the private, with the common charac-teristic of being non-profit. Cooperatives with employees and no aim for profit are part of this group. Finally, the informal refers to family, rela-tives, neighbours and the like. In comparisons of welfare systems in dif-ferent countries, it is common to look at a welfare mix constituted by the state, the market and the family. However, the voluntary sector is often neglected and unrepresented in such comparisons (Dahlberg, 2005; Graefe, 2004), although the original idea behind welfare pluralism was used in discussions concerning an increased engagement from the volun-tary sector (Dahlberg, 2005; Johnson, 1987).

User cooperatives, or consumer cooperatives, constitute one form of e.g. geriatric care, where the members consist of employees, the elderly, their relatives and others. These cooperatives are often firmly established and have a strong relation to the local community in which they are usu-ally located. In this sense they relate to the idea of welfare pluralism with a strong community association. The three geriatric care cooperatives included in the studies are examples of this, since they all started as a result of cooperation within a local context and with a strong relation to the community.

The fourth actor mentioned in welfare pluralism is the informal sector, i.e. family, relatives and the like. In the provision of geriatric care, rela-tives are increasingly often mentioned as a potential resource (SOU 1999:66), and in a comparison of the provision of home help between 1988/89 and 2002/03, there was a clear displacement from formal care to informal care provided by relatives (Larsson, 2006).

Changes in organisational characteristics As described previously, the development of a changing public sector has had large implications on organisational characteristics. Central themes in new public management that are related to organisational dimensions are decentralisation of responsibility from the statutory level down to the municipalities, increased individualisation, increased focus on the cus-tomer and increased use of performance measurements. With an empha-

••

9

sis on the market and competition, the situation is characterised by neo-liberal ideals, which have been more or less accepted and used in prac-tice. Parallel to new ways of organising, managing and putting emphasis on different control strategies, traditional control mechanisms, such as centralisation and formalisation, are still used. Research has shown how management can control its employees by either having a centralised structure, where the decisions are taken at the top of the hierarchy, or by a high degree of formalisation and standardisation, which limits the man-ager’s discretion (see e.g. Child, 1973; Mintzberg, 1979; Szulkin, 1996). These management ideals, inspired by the private sector and implemented by the public sector, also affect private and cooperative actors who have to adjust to the arena that is governed by the public, in the contract that is agreed upon by the separate parties. Along with these management ideals there are also certain laws and regulations associated with welfare work, which the providers of e.g. geriatric care, regardless of ownership form, have to follow. This illustrates a kind of reciprocity, where on the one hand public institutions have imitated the private sector as a result of aims for efficiency in combination with ideological change. On the other hand, alternative producers of e.g. geriatric care, regardless of ownership form, have to adjust to and thereby imitate the public sector, since this is where the prerequisites that are present for all providers of geriatric care are formulated and decided.

As a consequence of market-influenced ideologies, one of the most characteristic features within New Public Management is the emphasis on the customer as the main actor (Du Gay & Salaman, 1992; Forssell & Jansson, 2000; Rasmussen, 2004). A problem in connection with imple-menting ideals and ideas from the private market into public welfare is that pure business logic cannot be translated into welfare production without taking into account the specific characteristics associated with welfare services (Gustafsson, 2000). If welfare production is considered as just any service production this can have a negative impact on the em-ployees, who are educated and trained to see and take care of the needs of the elderly, but due to limited resources are prevented from conducting the work tasks that they see ought to be done (Rasmussen, 2004). In a study by Gustafsson & Szehebely (2005), results in that direction were presented, where the employees closest to the elderly felt inadequate since they could not provide enough help and support due to lack of re-sources.

10

Individualisation of work along with loyalty and commitment are cen-tral ideas behind Human Resource Management (Legge, 1995), and in Sweden as in other western countries the labour market has moved to-wards a situation where responsibility is increasingly placed on the indi-vidual, who expects to take responsibility for his/her own work situation (Allvin, 2004). Along with this decentralised responsibility it has become common within the public sector to empower employees to take their own decisions and act more autonomously within the frames that are given from above (Rasmussen, 2004; Wallenberg, 1997), while at the same time there has been a centralisation of strategic decisions (Hoggett, 1996; Rasmussen, 2004). With increasing influence from the private market, performance measurements such as key indicators and develop-ment talks to measure the results of employees, are becoming more common as a form of control measure (Dahlberg, 2005).

The development described above, with a changing public sector and welfare services under transition, also has consequences for the prerequi-sites for management. Studies on public welfare in Sweden have shown how first-line managers often have up to 90 employees to be responsible for (Andersson-Felé, 2003), while a majority of public managers consider around 30 employees to be a maximum number in order to perform a good job as a manager (SKTF, 2002). This development is a result of the decentralisation of responsibility for personnel and the operative work parallel to a centralisation of strategic decisions and resources (Rasmus-sen, 2004). The organisational prerequisites that are present for the man-ager contribute to the discretion and space that the manager has in order to carry out the work. Limited discretion affects the behaviour of the manager, and this in turn has implications for the psychosocial working conditions of the employees (Waldenström & Härenstam, 2006).

Psychosocial working conditions Increased demands for efficiency, limited resources and a decentralisation of responsibility are aspects related to the deregulation of the public sec-tor described earlier, that are often mentioned in explanatory terms in relation to psychosocial working conditions. In several studies of working conditions in the public welfare sector, a combination of increased re-sponsibility, decreased resources and increased demands has been de-scribed (see e.g. Wetterberg, 2000).

••

11

Worker control, or just control, is often used as a term to illustrate as-pects of influence, involvement and participation in the workplace. It is, however, important to define the concept, since there are parallel research traditions with different history and meaning. Control, or decision lati-tude, emanates from the field of occupational health and relates to being able to design jobs that are challenging for the single employee. It has been defined as “the working individual’s potential control over his tasks and his conduct during the working day” (Karasek, 1979:289), and there is thus a pronounced emphasis on the work situation for the single em-ployee. Decision latitude consists of two different but related constructs: skill discretion, capturing the certain qualifications and skills required for a job; and decision authority, concerning what to perform in the work and how to perform it. Although these two constructs are often used together as one measure, there is reason for treating them as two separate variables (Karasek, 1989). Social support concerns the degree to which an em-ployee can get support from supervisors and/or colleagues (Johnson, 1986) and it has been assessed in several different ways, e.g. as psycho-social climate, emphasising the relations between colleagues. Decision latitude, in combination with social support, has been found to have a buffering effect for high demands, as developed in the frequently used job demand–control model (Karasek & Theorell, 1990). Another meas-ure, mastery of work, relates to decision authority but measures the em-ployee’s perception of his/her own work performance and the quality of the work performed (Dallner, 2000).

Workplace participation in decision-making focuses on overall deci-sions and influence that are not just related to the individual’s work situa-tion but more to workplace democracy as a whole. The development of the concept goes back to the early 1930s, when Elton Mayo discovered the employee as a social nature (see e.g. Huzzard, 2003; Jacobsen & Thorsvik, 1998; Pugh & Hickson, 1993). In Scandinavia, there has been a long research tradition in the field of workplace democracy, especially within industrial production, with several important contributions from e.g. Dahlström (1969), Gardell (1981) and Thorsrud & Emery (1969). In the increased global competition that arose during the 1970s and 1980s, with an increased focus on service production with the customer in the centre, corporations began to see participation programmes as instru-ments for increasing efficiency (Smith, 2006). The bottom line of the assumption is basically that when workers are able to influence decisions concerning what they produce and how they produce it, they will become

12

more dedicated and thereby more productive (ibid). The growing interest of the customer is based on a neo-liberal assumption of the sovereign customer in a free market (Sturdy & Korczynski, 2005), and there are numerous management strategies including themes of participation and empowerment that have been practised, including e.g. Total Quality Management and Human Resource Management (Bolman & Deal, 1997; Huzzard, 2003). Some of these strategies have also gained ground in wel-fare. In an example from Norwegian home help for the elderly, referred to above, employees were empowered to act independently within given frames (Rasmussen, 2004). Parallel to the description above, where par-ticipation is used more as a means for employers to increase productivity, in value-based organisations like cooperatives participation is practised and used due to its own value (Cheney, 2002; Rothschild-Whitt, 1979). Here, workplace democracy and participation are the core values of the organisations rather than a temporary management strategy. However, in an increasingly competitive market, where also collective ideologist or-ganisations have to act, there might be difficulties in holding on to these core values, something that has been recognised in the Basque coopera-tives at Mondragón (Cheney, 2002). Participation and being able to influ-ence are also important features within Workplace Health Promotion, a research tradition with a salutogenic perspective, aiming at promoting health and developing positive and healthy aspects rather than curing ill health (Antonovsky, 1996).

Summary The aim of this section has been to illustrate the different characteristics and motives in public, private and cooperative organisations; the devel-opment of a changing arena for welfare services towards an increased number of private and cooperative actors; management control strategies and changes in organisational prerequisites, and certain characteristics of psychosocial working conditions for employers and employees. Given these circumstances it is reasonable to assume that there will be differ-ences in both organisational characteristics and psychosocial working conditions between different forms of ownership and where different forms of services are produced. The empirical results concerning psycho-social working conditions, management control strategies and prerequi-sites for management are presented in a later chapter, but before that, in the next chapter, we deal with method and study design.

••

13

Design, method and description of data The research work constituting this thesis is part of the Healthy Work-place Study (HWS), which in turn was developed from the WOLF study2. The overall aim of HWS was to explore the significance of organisational conditions and individual characteristics on the variation between organi-sations in working conditions and employees’ health (Härenstam et al., 2006). The section below describes the data and the design of the four included studies. Table 2 on page 18 gives a brief overview of these is-sues including chosen methods, something that is more thoroughly de-scribed and discussed in each of the attached articles.

Participants and procedure Our primary interest concerning the sample was at the organisational level, i.e. workplaces and not single individuals. Since it was a follow-up of an ongoing study, as mentioned in the footnote below, most of the workplaces were inherited from the previous data collection process. In order to achieve a sample that was as much in line with the structure of the Swedish labour market as possible, a complementary sample was collected. This resulted in a total of 90 workplaces, of which 30 were within manufacturing and 60 belonged to the service sector. For the stud-ies included in this thesis, only the 60 workplaces within service produc-tion were used. Our criteria for a workplace were that it had to have a separate address, a separate profit centre (result unit) and a separate man-ager (Härenstam et al., 2006). All workplaces were situated in mid-Sweden. In these workplaces, structured interviews were conducted with the local managers, concerning issues related to organisational character-istics such as the production process, market competition, ownership, organisational change, management strategies, work environment and HRM issues. There was no interest in subjective opinions and percep-tions, but rather objective measures concerning actual organisational pre-requisites. The interviews lasted on average 1½ hours and were con-

2 The aim of WOLF was to look for relations between work environment issues and risk

factors for coronary heart disease. The project was carried out in two counties in Sweden between 1996 and 1998, with a follow-up around five years later between 2000 and 2003. The follow-up included the Healthy Workplace Study, where organisational data was collected at the workplaces in which the individuals participating in WOLF worked.

14

ducted by two researchers: one with responsibility for the questioning and the other for writing and filling in the questionnaire. The data collection took place between 2000 and 2003. All participating workplaces received feedback in terms of a summary of the main results for their separate workplace in relation to the whole material.

In the selected workplaces all employees were invited to participate in the study by filling in an individual questionnaire concerning lifestyle issues such as smoking, drinking, exercise and eating habits, employment arrangements such as e.g. working hours, psychosocial and physical work environment, social relations, individual characteristics, socio-demo-graphics, sleeping habits and finally a part concerning physical health status. In a majority of the workplaces the Occupational Health Services, OHS, functioned as a field organisation and distributed and collected the questionnaires. When OHS was not involved, the researchers handled this process. The respondents were given the questionnaire at work with a reply envelope, so that they could fill it in at home.

Already in the first study it was clear that the employees working in the care sector should be treated separately, since the working conditions there are special, and also because all the cooperatives were found there. In article I, page 258 (attached), there is a table summarising background data for the included workplaces and employees.

The data for the first three studies was collected within the same pe-riod and described in the introduction to this chapter. For the fourth study, based on separate interviews with seven geriatric care managers, complementary data was collected between March and May 2006. The managers, as well as their employees, had participated in the earlier, lar-ger project, which made it easier for them to participate in a second inter-view. Four of the managers were still in the same job and three of them had started in the position after the former data collection. The seven managers were chosen due to their formal placement in the organisation, i.e. as first-line managers at the lowest level above the work group. The interviews, which lasted on average 1½ hours, were semi-structured and contained question areas concerning prerequisites for management.

In the whole material, five cooperatives and 52 employees were in-cluded. Three of these are worker cooperatives, i.e. they are owned and run by the employees who hold membership. The two others are con-sumer cooperatives, where the members consist of a combination of em-ployees, users (the elderly and their relatives), neighbours and others. Due to the rather small number of cooperative employees, no distinction

••

15

was made between members and non-members. In the three geriatric care cooperatives included in study III, only 12% were members, whereas in the other two, which were both pure worker cooperatives, almost all em-ployees were members. The basic idea in cooperatives is that influence and participation is associated with membership. Here, however, the members and non-members have been treated equally, based on the idea that many of the cooperative ideals associated with taking an active part in decision-making would be beneficial also for non-members, and would contribute to cooperative values in the workplace at large. With a larger sample, it would have been possible to distinguish between members and non-members.

Design of the thesis and the included studies With the research focus on the relationship between different forms of ownership and psychosocial working conditions, the four separate studies included in the thesis were designed to go from the all-embracing and general in the first study, to the detailed and more narrowly investigated issues raised in the fourth study. In the first study, the aim was to seek an answer to the question of whether there are reasons to believe that differ-ent kinds of ownership might result in differences in perceived demands, control and psychosocial climate. In all, 1384 employees in 60 work-places participated, and this constituted the whole data set. Since the ownership forms are not equally distributed over different types of opera-tions (e.g. all cooperatives are within the care sector), this aspect had to be taken into account when making comparisons. We thus added type of operation into the analyses.

The second study focused on management control strategies with rele-vance for psychosocial working conditions, and whether there are any differences in these aspects due to ownership form. Of the total of 60 workplaces (used in study 1) 55 were included. The five workplaces that were public enterprises were removed, since that ownership form was not of primary interest for the study. Here only organisational data, collected from interviews with the managers, were used – no individuals were in-cluded. Eight management control strategies were chosen, theoretically related to ownership, type of service and to psychosocial working condi-tions. Since the types of operations were fairly segregated in the different ownership forms, as mentioned above, we chose to take this into account

16

by also making separate analyses comparing care organisations with ser-vice organisations in general for potential differences in the organisa-tional dimensions. The twelve care organisations consisted of e.g. geriat-ric care, a health care centre and a hospital care department.

Based on the results in the previous studies, showing the importance of type of operation and the specific requirements associated with caring but still focusing the main interest on making comparisons between own-ership forms, the third study was designed according to these prerequi-sites. Accordingly, the focus was on 152 employees working in seven geriatric care organisations (of the total of 60) in three different forms of ownership. All workplaces were about the same size (although the coop-eratives had fewer employees than the others), with the same type of pro-file, and where all were under the surveillance of the local municipality. This was a way of coming closer to the heart of the matter and answering the central and main question in the thesis, i.e. whether there are differ-ences in psychosocial working conditions in different ownership forms. More precisely, the research question concerned differences in perceived control and participation.

Finally, the fourth study focused on a related aspect of psychosocial working conditions, i.e. prerequisites for management in different forms of ownership. Here the managers of the seven geriatric care institutions (in which the employees included in study III worked) were interviewed on organisational prerequisites for management and the managers’ discre-tion.

The use of data for the separate studies is illustrated in the following figure.

••

17

Figure 2. Illustration of how the data has been used in the separate studies (the whole data set is used in study 1 and then gradually reduced.

I 60 workplaces with 1384 employees

(the whole service sector)

II 55 workplaces

III

7 workplaces within geriatric care with 152 employees

IV

7 geriatric care first-line managers

As described above, the whole dataset consisted of data from the individ-ual questionnaire to 1384 individuals, and interviews with the managers in the 60 organisations in which they work. This material was used for the first study. In the second study, data on 55 of the 60 workplaces were included but with no individuals. In the third study only seven of the workplaces, with 152 employees and all within caring, were used. For the fourth study complementary data was collected, including interviews with the seven managers of the workplaces in study 3. The overall structure and design of the studies is illustrated in figure 3 and in table 2, presented below.

18

Tabl

e 2.

Ove

rvie

w o

f the

incl

uded

stud

ies .

St

udy

Res

earc

h qu

estio

n In

depe

nden

t var

iabl

e D

epen

dent

var

iabl

e D

ata

Met

hod

I A

re th

ere

any

diff

eren

ces

in p

erce

ived

psy

chos

ocia

l w

orki

ng c

ondi

tions

for

empl

oyee

s wor

king

in

publ

ic, p

rivat

e,

coop

erat

ive

or p

ublic

en

terp

rise

wor

kpla

ces?

Ow

ners

hip ,

i.e.

pub

lic,

publ

ic e

nter

pris

es,

coop

erat

ive,

pri

vate

. Ty

pe o

f ser

vice

, i.e

. kn

owle

dge-

inte

nsiv

e se

rvic

e, la

bour

-inte

nsiv

e se

rvic

e, c

arin

g, te

achi

ng.

Psyc

hoso

cial

wor

king

co

nditi

ons i

n te

rms o

f de

man

ds, c

ontr

ol,

psyc

hoso

cial

clim

ate.

Que

stio

nnai

re to

138

4 em

ploy

ees i

n 60

wor

kpla

ces:

25

pub

lic w

ith 7

96

empl

oyee

s; 2

5 pr

ivat

e w

ith

431

empl

oyee

s; 5

coo

ps w

ith

52 e

mpl

oyee

s and

5 p

ublic

en

terp

rises

with

105

em

ploy

ees.

Mul

tilev

el a

naly

sis f

or

sepa

ratin

g th

e va

rianc

e in

w

orki

ng c

ondi

tions

be

twee

n di

ffer

ent l

evel

s, an

d fo

r pre

dict

ing

the

impo

rtanc

e of

ow

ners

hip

and

type

of s

ervi

ce.

II

Are

ther

e an

y di

ffer

ence

s in

the

pres

ence

of

man

agem

ent c

ontro

l st

rate

gies

in d

iffer

ent

form

s of o

wne

rshi

p, a

nd

seco

ndly

in d

iffer

ent

type

s of s

ervi

ces?

Ow

ners

hip ,

i.e.

pub

lic,

coop

erat

ive,

pri

vate

. Ty

pe o

f ser

vice

, i.e

. non

-ca

ring

ver

sus c

arin

g.

Man

agem

ent c

ontro

l st

rate

gies

in te

rms o

f ce

ntra

lisat

ion,

fo

rmal

isat

ion,

gro

up- a

nd

indi

vidu

al re

spon

sibi

lity,

ha

rd a

nd so

ft pe

rfor

man

ce

cont

rol,

cust

omer

ad

just

men

t and

cus

tom

er

cont

rol.

Inte

rvie

ws w

ith m

anag

ers i

n 55

wor

kpla

ces w

ithin

se

rvic

e: 2

5 pu

blic

, 25

priv

ate

and

5 co

oper

ativ

e.

Stru

ctur

ed in

terv

iew

s, w

hich

wer

e pr

oces

sed

stat

istic

ally

for m

ean

com

paris

ons b

etw

een

owne

rshi

p fo

rms a

nd

betw

een

wel

fare

and

se

rvic

es in

gen

eral

. K

rusk

al-W

allis

and

Man

n-W

hitn

ey.

III

Are

ther

e an

y di

ffer

ence

s in

how

em

ploy

ees i

n ge

riatri

c ca

re in

diff

eren

t fo

rms o

f ow

ners

hip

perc

eive

thei

r con

trol a

nd

parti

cipa

tion?

Ow

ners

hip,

i.e.

pub

lic,

coop

erat

ive

and

priv

ate.

Ps

ycho

soci

al w

orki

ng

cond

ition

s, in

term

s of

cont

rol (

deci

sion

aut

horit

y an

d m

aste

ry o

f wor

k) a

nd

part

icip

atio

n.

Que

stio

nnai

re to

152

em

ploy

ees w

ithin

7 g

eria

tric

care

inst

itutio

ns: 3

2+47

in

two

publ

ic; 9

+7+1

8 in

thre

e co

oper

ativ

e an

d 15

+24

in

two

priv

ate.

Mea

n co

mpa

rison

s of

cont

rol a

nd p

artic

ipat

ion

in

the

owne

rshi

p ca

tego

ries.

Kru

skal

-Wal

lis a

nd

AN

OV

A.

IV

Wha

t are

the

prer

equi

site

s fo

r man

agem

ent i

n ge

riatri

c ca

re in

stitu

tions

w

ith d

iffer

ent f

orm

s of

owne

rshi

p?

In th

is q

ualit

ativ

e st

udy

ther

e w

as n

o gr

oupi

ng o

f the

dat

a du

e to

ow

ners

hip

form

prio

r to

anal

ysis

. In

the

anal

ysis

ce

rtain

sim

ilarit

ies w

ithin

an

owne

rshi

p fo

rm w

ere

iden

tifie

d bu

t als

o de

pend

ing

on w

hat m

unic

ipal

ity th

e or

gani

satio

n be

long

ed to

. The

ow

ners

hip

form

s wer

e pu

blic

, coo

pera

tive

and

priv

ate.

The

pre

requ

isite

s for

m

anag

emen

t con

cern

ed o

rgan

isat

iona

l cha

ract

eris

tics,

resp

onsi

bilit

y, d

eman

ds, r

esou

rces

and

are

na fo

r di

alog

ue. M

anag

ers’

dis

cret

ion

in te

rms o

f the

ir ow

n pe

rcep

tion

was

als

o ex

plor

ed.

Inte

rvie

ws w

ith 7

man

ager

s of

7 g

eria

tric

care

inst

itutio

ns

(the

inst

itutio

ns in

whi

ch th

e em

ploy

ees i

n th

e pr

evio

us

stud

y w

orke

d).

Sem

i-stru

ctur

ed,

trans

crib

ed in

terv

iew

s. A

m

odifi

ed v

ersi

on o

f co

nten

t ana

lysi

s, se

ekin

g fo

r pat

tern

s of r

egul

ariti

es

in th

e w

hole

, bet

wee

n ow

ners

hip

form

s and

be

twee

n di

ffer

ent

mun

icip

aliti

es.

••

19

Prerequisites for management and manager’s discretion

The model in figure 1 on page 3, summarised the relationship between the main theoretical perspectives in the thesis. The figure below, with several similarities to figure 1, shows how the four included studies con-stitute parts of the whole. Figure 3. The relationship between the four studies included in the thesis.

DETERMINANTS ORGANISATIONAL PSYCHOSOCIAL CHARACTERISTICS WORKING CONDITIONS

Contextual factors Organisational Individual level level

The independent variables ‘ownership’ and ‘type of service’ can be seen as contextual, solid factors that can be considered as constant. Data have been collected at the organisational level to measure management control strategies and prerequisites for management, but the manager’s discre-tion, which is a consequence of the prerequisites for management and here considered as a psychosocial working condition for the single man-ager, is measured at the individual level. The relationships between the included dimensions are explored in four separate studies, which are at-tached in this thesis.

IV

II

Ownership Type of

service (fo-cus on wel-

fare)

Management control strategies

Demands, control and psychosocial

climate I

Control and participation III

20

In study I, demands, control and psychosocial climate as perceived by the employees constitute the dependent variables, and in study III a com-parison of how the employees perceive their control and participation is made. The outcomes in these studies are thus based on individual data from the questionnaire. Study II focuses on management control strate-gies (based on the interviews) in relation to different forms of ownership, and finally study IV, based on complementary interviews, concerns the prerequisites for management and manager’s discretion within geriatric care with different forms of ownership. As shown in the box above, the prerequisites for management and manager’s discretion are related and woven together, and not separated in the qualitative analysis. Compared with figure 1 (page 3), the aim of this figure is to show the relationship between the investigated variables and between the included studies, and therefore the two dashed arrows in figure 1, illustrating a possible theo-retical association, have been removed here. A more detailed description of e.g. the included variables is presented in each of the articles.

••

21

Empirical findings In the section below, the main findings from the four included studies are presented. The chapter is structured in relation to the outcomes in the separate studies. Consequently, studies I and III are described under the heading “Psychosocial working conditions”; study II, dealing with organ-isational characteristics, is described under “Management control strate-gies”, and finally study IV is described under “Prerequisites for manage-ment”.

Psychosocial working conditions (study I and III) The initial question in the first study dealt with the degree to which fac-tors at different organisational levels affect psychosocial working condi-tions for the individual employee. The psychosocial working conditions measured were demands, control (decision latitude) and psychosocial climate. As a complement to other studies comparing private and public ownership in relation to psychosocial working conditions, this one com-pared four forms of ownership (public, public enterprises, cooperatives and private enterprises) in relation to the outcomes. A second aim was to investigate potential differences in these psychosocial working conditions for different forms of ownership and different types of services. The data consisted of 1384 individuals working in 60 workplaces belonging to 25 establishments. Four hypotheses were formulated. The first one suggested that there is a systematic variation in how employees consider their psy-chosocial working conditions that is attributed to different organisational levels. The results of multilevel analyses showed that a rather high pro-portion of variance (13% for demands, 19% for control and 10% for psy-chosocial climate) was above the individual level, thus indicating that aspects at an organisational level were relevant to highlight, and that mul-tilevel analysis would be appropriate as a method. This result supported the first hypothesis. The second hypothesis suggested that employees working in cooperatives consider their control, i.e. decision latitude, to be greater and their psychosocial climate to be better compared with those employed in other ownership forms. There were no significant results supporting that the control would be higher in cooperatives, although there were tendencies in that direction. Instead, employees working in public enterprises perceived significantly lower control than those em-

22

ployed in the public sector, which was the reference category. Employees in cooperatives perceived a better psychosocial climate than those in the public sector, supporting part of the second hypothesis.

Type of service was also an important aspect in relation to control. In line with the third hypothesis, the lowest control was perceived for those working within caring and also for employees within labour-intensive services. The fourth hypothesis, suggesting that the highest demands were perceived among those working within human services, i.e. caring and teaching, was supported. The work for employees within caring is charac-terised by both high demands and low control. Since the types of services in the ownership forms are not equally distributed (e.g. all included coop-eratives are within caring) it seemed relevant to take this aspect into ac-count for the following studies.

The main idea behind study III was that employee control and partici-pation can be seen as two related but distinctly different aspects of per-ceived influence. Whereas the well-established measure of decision au-thority (as one part of the decision latitude, or control index) measures the degree to which the employee can influence aspects of his/her own work situation, participation refers to an overall influence over decisions re-lated to the workplace as a whole and not just the individual work tasks. In all, 152 employees within caring in public, private and cooperative form participated. Of these, 79 belonged to two public organisations; 34 belonged to three cooperatives and 39 were from two private organisa-tions. One hypothesis assumed that employees in cooperatives would perceive higher participation than those in private and public organisa-tions, since participation and influence are strong values in the coopera-tive tradition. For control it was assumed that there would be no differ-ences between the ownership forms, since the provision of care is so re-stricted by rules, regulations and legislation, regardless of ownership form. Participation was captured by four different variables / indexes and in two of these, voice–response concerning work environment issues and participation in overall decision-making, those employed in cooperatives perceived higher participation. In another question concerning whether the manager gave adequate information (which was assumed to be a pre-requisite for participation and therefore a relevant measure), higher val-ues were shown for those employed in public geriatric care. In the fourth measure, concerning the degree to which the employees could participate in decisions on recruitment of colleagues and managers, those employed in private geriatric care showed higher values.

••

23

For measuring control, two well-established indexes were used: one of decision authority, i.e. the influence over what to perform in the work and how to perform it; and the other of mastery at work, i.e. the employee’s perception of her/his own work performance and the quality of the per-formed work. Mean rank comparisons between the three groups of em-ployees showed that in both these indexes there were no significant dif-ferences due to ownership form, thus supporting the hypothesis and the view that individual opportunities to influence work content in geriatric care are rather limited. A valuable conclusion from this study was that the sometimes carelessly used concept of control as synonymous with par-ticipation here showed that it captures entirely different aspects – indi-vidual control over the own work situation versus influence over and participation in decision-making concerning the work place as a whole – and that participation but not control was decisive when comparing per-ceptions from public, cooperative and private geriatric care employees.

Management control strategies (study II) The aim of the second study was to compare how four types of control strategies were manifested in organisations with different forms of own-ership and in care versus non-care organisations. The data consisted of structured interviews with the local managers at 55 workplaces: 25 pub-lic, 25 private and 5 cooperative. Twelve of the organisations were within the care sector, e.g. geriatric care and a health care centre; the rest be-longed to service as a whole and included a wide range of organisations in e.g. IT-support, system design, libraries, bakery and teaching. The control strategies of interest were categorised into four different types and eight measures: by the formal structure (centralisation and formalisation), by placement of responsibility for planning and follow-up (on the indi-vidual or the group), by output control (hard or soft performance control) and finally by indirectly placing the control on the customer (customer adjustment and customer control). Indexes were constructed from sepa-rate questions and mean rank comparisons were made between the differ-ent forms of ownership and between organisations within caring and ser-vice in general. There were significant differences between ownership forms in four of the eight measured strategies: centralisation, group re-sponsibility, individual responsibility and quantitative performance con-trol. Private enterprises were more centralised than public and coopera-

24

tive ones, the use of placing the responsibility on the group was more pronounced in both cooperatives and public organisations than in the private ones, and the use of placing the responsibility on the individual was more pronounced in the public sector compared with the other two. In the private organisations it was also more common to use quantitative performance control than in cooperatives and the public sector.

Although this was not significant, the cooperatives were more formal-ised than the other ownership forms. The questions constituting the index covered issues typically prevailing within welfare services, e.g. rules, legislation and standard procedures associated with the provision of care. This illustrates the importance of taking the kind of service being pro-duced into account when interpreting the results.

Differences in the use of control strategies were also found in a com-parison between the service sector in general and the organisations pro-viding care. The care organisations were more formalised than the service sector in general and they also had a more pronounced use of group re-sponsibility. The service sector at large used qualitative performance control to a greater extent than the care organisations. When looking at the separate questions constituting the indexes (and presented in the ap-pendix in the article), the care organisations were characterised by a high degree of formalisation, performance control in terms of result measure-ment, e.g. key indicators, and a strong influence from customer control. These are aspects that might indirectly have a negative impact on the working conditions for employees working in the care sector.

Prerequisites for management (study IV) The idea of the fourth study, as well as the second one, was to investigate and describe some organisational characteristics that are of indirect im-portance for working conditions. More precisely, the aim was to increase the understanding of how the prerequisites for management are mani-fested in public, cooperative and private geriatric care. Interviews were conducted with seven line managers in geriatric care: 2 public, 3 coopera-tive and 2 private, and all seven in the same formal position, i.e. they constituted the lowest level of management, directly above their work group. The prerequisites for management concerned the organisational settings, demands, responsibility, resources and arena for dialogue, i.e. being able to communicate with the administrative and political level, but

••

25

also with the employees. There was no intention of grouping the inter-views in advance and analysing them in relation to ownership form – all seven interviews were read several times separately and treated individu-ally. The second step involved a search for common patterns but also for differences due to ownership or municipality belonging.

The results showed similarities in the organisational structure between the public and the private. Another organisational level had been incorpo-rated in both cases but with entirely different effects for the first-line managers included in the study. In the public organisations another level was added above the responding managers, with the result that they ended up further away from strategic planning and decision-making. For the private managers, on the other hand, the new level was added below the managers, giving them more time and space to deal with strategic issues and with a small number of three group leaders instead of having the daily, operative responsibility for 35-50 employees. The cooperatives are different with a fairly autonomous and demarcated type of organisa-tion, where the organisation literally starts and ends with the building.

Administrative issues to a high degree characterise the work content for all managers. The public managers had a demanding responsibility for around 35-50 employees with whom they had to hold both development talks and separate salary talks as well. The fact that they had such large work groups affected their work situation, where much of the responsibil-ity was concerned with personnel issues. In addition, they also had the responsibility for financial issues although one of them was not given the opportunity to be involved in the budget process. As mentioned, the pri-vate managers focused much more on strategic issues that to a high de-gree concerned matters of finance, and because the group leaders handled all operative personnel tasks, they only had to deal with the more extra-ordinary and difficult personnel issues. They had only three persons (the group leaders) to have development talks with.

In two of the cooperatives the responsibility for personnel had been given to someone other than the line manager and this created some insu-larity concerning the responsibility for different issues. All seven manag-ers considered that the demands were both high and complex, and that the financial demands for keeping within the budget were not negotiable but just something to accept. The cooperative line managers considered that the external demands from e.g. the National Food Administration, the Swedish Work Environment Authority and the county administrative board, were difficult to cope with and they wanted some kind of support

26

from a mentor or a coach. The public managers also called for an extra resource, whereas the private ones felt no such need.

Besides the common aspects of being in the same context, i.e. geriatric care, there were some other similarities between the prerequisites for the seven line managers that have already been mentioned: large demands concerning administrative tasks and no opportunity to negotiate the budget. However, within these certain, predetermined circumstances, the way in which the organisations in private care had been structured, by implementing another managerial level below, resulted in them coming closer to strategic decisions and having more influence over the financial planning and follow-up of results. For the public managers, on the other hand, where another organisational level was implemented above them, they still had to deal with both large amounts of administrative tasks con-cerning the large work groups, holding development talks and salary talks, and still had the responsibility for the budget and follow-up without being able to influence. For the cooperative managers, who had rather heterogeneous prerequisites, a common pattern was the desire for an extra resource in terms of a coach in order to deal with the divergent demands.

When comparing the work situation for the seven managers included in this study, the private line-managers seemed to have the best prerequi-sites in their position to perform a good job. They could focus on finan-cial and strategic issues, which made their work more delimited and al-lowed them more discretion.

••

27

Discussion The overall aim of this thesis was to investigate the relationship between different ownership forms and psychosocial working conditions. In addi-tion to this, a second aim was to investigate whether or not differences in ownership form also created different organisational characteristics (in terms of management control strategies and prerequisites for manage-ment, which are relevant aspects in relation to psychosocial working con-ditions). In this final chapter the main results of the included studies are discussed. Employees in cooperatives considered their psychosocial climate to be better than those working in the public sector (study I). The index cov-ered aspects related to the relationship with and support from colleagues, with a collective focus. The organisational data (study II) also showed that cooperatives were characterised by placing responsibility for plan-ning and follow-up of results on the group. In one of the measures of participation (study III) in terms of overall decision-making and voice–response concerning work environment issues, employees in cooperatives showed higher values than those in the public and private sectors. It is possible that employees who perceive support from colleagues, and who are joined together by group organisation and group responsibility, also consider that it is easier to express their opinion when they experience support from colleagues in a value-based organisation like cooperatives (Cheney, 2002; Rothschild-Whitt, 1979). It is also in accordance with the organisational goals in cooperatives as primarily social (Nilsson, 1986; Sjöstrand, 1993). No differences in the measures of control were found between the ownership forms (study III). This variable relates more to whether it is possible for individuals to influence their own work situa-tion, and it is reasonable to assume that for employees in geriatric care, where tight financial frames and limited resources characterise the work (study IV), there is no room for individual solutions and adjustments of the work content. One might raise the question of whether the market situation of which the included cooperatives are part, competing with other providers of geriatric care, creates a situation that limits some of the genuine cooperative ideals. As described in the theoretical part, there might be difficulties in holding on to core values and workplace democ-racy in an increasingly competitive market (Cheney, 2002).

28

Employees’ access to information is one structural factor that is neces-sary in an organisational empowerment environment, together with sup-port, resources and opportunities in the work setting (Kanter, 1993). In study III, information about central workplace issues related to organisa-tional goals was considered a prerequisite for active participation. How-ever, having information about central issues without sufficient resources to deal with the information can create a burden for the employee. It is therefore important to consider the context in which the information is given. Public care employees considered that their managers were better at giving information than those in cooperatives and private geriatric care. According to study IV, the financial resources within geriatric care were rather limited, and this can create a negative situation for the employees as well as for the managers. Therefore, the manager has to balance the information she gives with the potential resources available. For the same variable concerning information, employees in cooperatives considered that their managers did not give adequate information and thereby were not supportive in facilitating participation. In study IV, where the manag-ers of the employees in study III were interviewed, it was clear that for two of the three cooperative managers, the responsibility for finance and personnel was divided between two persons, which created some confu-sion concerning the roles. This situation could perhaps also create a gap resulting in information “falling between two stools”. Another explana-tion, which was discussed in study III, could be that in cooperatives, with an emphasis on collective values, individual leading roles might not be encouraged, and giving and receiving information might be seen as a common responsibility for all employees in the workplace – not just a responsibility for the manager. From the theoretical description of charac-teristics of ownership forms (page 4), the type of communication was assumed to be vertical in the public sector, and complex and mainly hori-zontal in the cooperatives (Nilsson, 1986; Sjöstrand, 1993). A fourth pos-sible explanation for the differences between the public and cooperative organisations is related to resources. In one-way communication, which aims at dropping information downwards, the time and effort involved is moderate. In cooperatives, on the other hand, with their demands for a democratic procedure, the process is much more time-consuming, or as cited by Rothschild-Whitt (1979:519): “Democracy takes time. This is one of its major social costs. / Quite simply, a boss can hand down a bu-reaucratic order in a fraction of the time it would take a group to decide the issue democratically”. Considering the divergent and complex work

••

29

situation with an imbalance between demands and resources as described by the cooperative managers (study IV), it is plausible to assume that time-consuming meetings for giving information might not always be of priority. In order to save time, the character of the types of meetings in which information is given and decisions are to be taken is thus of great importance. According to Rothschild-Whitt (ibid), a combination of planning, practice and discipline, in combination with a reasonable amount of time available, can increase the potential for democracy and participation in the workplace.

In relation to the result that employees in cooperatives considered lack of information from their managers, there appears to be a contradiction to the fact that the same employees perceived better participation in terms of being able to influence decisions concerning the workplace as a whole (study III). Their cooperative managers also perceived good relations and connections with the board (study IV). Two of the three managers were members of the board themselves, and the third manager perceived that she had good contacts with members of the board and that she was lis-tened to. It is reasonable to assume that these are prerequisites that con-tribute to facilitating for employee participation. Here it seems as if the employees still believed they had better opportunities to influence deci-sions in the workplace as a whole, even though their cooperative manag-ers did not give them adequate information. Presumably, the decisions in which the employees participated concerned other factors that did not demand information in advance from the manager.

The managers in the two private geriatric care institutions appeared to have better prerequisites for management compared with the public and cooperative managers (study IV). The primary reason for this difference was the way in which they had been reorganised around three years ago, resulting in them now having three group leaders below them to take the responsibility for daily, operative work tasks concerning employees, the elderly and their relatives. This change facilitated for the managers and gave them more time for strategic work, primarily financial planning and follow-up of the budget, which in combination with what was described as almost unlimited resources for support within the organisation, re-sulted in a good work situation for the line managers. The individual level data, i.e. the questionnaires to all employees, were collected between 2000 and 2003, and this was thus done before the reorganisation was carried out. How the employees, including the group leaders, perceive their working conditions after this change would be interesting to follow,

30

especially to compare potential differences in relation to the situation before the organisational change. In article IV, there is further reading concerning these results.

The type of service produced turned out to be of central importance when comparing psychosocial working conditions. Organisations provid-ing care were characterised by a variety of aspects: they were highly for-malised, with a pronounced group organisation and frequent use of quan-titative performance control, and there was also a high degree of interac-tion with the customer and customer control (study II). Further, they were characterised by tight financial resources that were not negotiable, a great deal of administrative tasks and high demands (study IV) and in addition the employees working within the care sector also perceived high de-mands and low control (study I). It was also mentioned in the interviews in study IV that there was a fairly large degree of freedom within the given framework, illustrating a decentralisation of operative responsibil-ity, while decisions in terms of resources are taken by top management. However, when all types of ownership are ordered under the public re-sponsibility and are in a dependent position (because they want the chance to win a contract for the provision of geriatric care), then many of the differences disappear, e.g. there are no differences in perceived con-trol (study III). This can be interpreted in such a way that statutory gov-ernance sets the limits and the frames and, in combination with the spe-cific rules, regulations and legislation associated with geriatric care, there are only limited opportunities to influence working conditions.

According to the discussion above, and in relation to the questions raised in the introduction, there are differences between ownership forms with regard to psychosocial working conditions, management control strategies and prerequisites for management, and these differences occur in all of the included studies. However, when the comparisons were re-stricted to only one type of operation, i.e. care, it is rather the certain similarities within care organisations (and geriatric care), regardless of ownership form, that are most pronounced.

When returning to the theoretical figure that has been the basis of this work, presented on page 3, it might be relevant to make a small revision of the model. Instead of having one box for both ownership and type of service, where type of service was seen more as a sub-ordered variable to ownership, these two variables should rather be separated and given one box each, since the type of service, i.e. caring, turned out to be even more important than the ownership form.

••

31

In December 2006 one of the cooperatives included in the studies had to give up competing for a contract with the municipality concerning the provision of elderly care, which they had been carrying out since the start of the cooperative around ten years ago. In the last contract process they lost the tender and by January 2007 another large and private care com-pany had taken over the running of the elderly care. Research has shown how competition and threat of change in operational form (principal re-sponsibility) create anxiety and frustration among the employees (Andersson, 2002), and also how small enterprises and cooperatives have difficulties in the procurement process compared with the large, private firms, when it comes to knowledge of law and how to submit a tender (Cantzler, 2000). In that sense there is no competition on equal terms. The entrepreneurs who start the cooperatives have created their own jobs and are therefore able to work where they live, in these cases in small villages in the countryside, instead of having to commute long distances to work. These are aspects that, although not included in these studies, can be related to working conditions in a broader perspective, perhaps including quality-of-life issues. This kind of systematic study, where relationships in all parts of the distribution chain are investigated, is rather rare and has been neglected by researchers (Wallenberg, 2000). Such a study could play a valuable role when making decisions concern-ing the provision of future care for the elderly and the working conditions of the employees. The separate studies in this thesis have concentrated on the psychosocial working conditions that obviously appear at work, and the questions measuring working conditions are of the kind that are re-lated to the workplace. For future studies it would be valuable to conduct a study with a more comprehensive research design, including not only working conditions, organisational dimensions and prerequisites for man-agement in comparisons between ownership forms, but also including the quality of care from both the clients’ and other stakeholders’ point of view.

In order to answer the questions raised in the thesis, different methods have been used, based on a rich material consisting of both data from single individuals and organisational data from interviews in 60 work-places. In addition, complementary data has been collected, resulting in material from seven interviews being used in one qualitative study. There has also been an aim to try to connect typical individual-level issues con-cerning psychosocial working conditions with issues concerning organ-isational characteristics, where individuals are often excluded and not

32

taken into account. Hopefully, some of the results can contribute to the main question concerning the relationship between organisational charac-teristics and psychosocial working conditions in different forms of own-ership, especially in the provision of welfare services.

••

33

References Allvin, M. (2004) The Individualization of Labour. In Garsten, C. &

Jacobsson, K. (eds) Learning to be Employable. New agendas on work, responsibility and learning in a globalizing world. Basingstoke. Palgrave Macmillan.

Andersson, F. (2002) Konkurrens på kommunala villkor. Om konkur-

rensutsättning och gränsen mellan marknad och byråkrati. Stock-holm. Svenska Kommunförbundet.

Andersson-Felé, L. (2003) Hur många underställda kan en chef ha? Om

kontrollspann i vård och omsorg. Stockholm. Socialstyrelsen. Antonovsky, A. (1996) The salutogenic model as a theory to guide health

promotion. Health Promotion International, 1 (1): 11-18. Bolman, L.G. & Deal, T.E. (1997) Nya perspektiv på organisation och

ledarskap. Lund. Studentlitteratur. Borell, K. & Johansson, R. (1998) Den nätverksbaserade nätverksmodel-

len - exemplet Ädel. In Lindqvist, R. (ed) Organisation och välfärds-stat. Lund. Studentlitteratur.

Cantzler, I. (2000) I privat regi - att bedriva vård och omsorg i nya for-mer. Stockholm. NUTEK.

Cheney, G. (2002) Values at Work. Employee Participation meets Market

Pressure at Mondragón. Ithaca. Cornell University Press. Child, J. (1973) Predicting and understanding organisation structure. Ad-

ministrative Science Quarterly, 18: 168-185. Dahlberg, L. (2005) Interaction between Voluntary and Statutory Social

Service Provision in Sweden: A matter of Welfare Pluralism, Substitu-tion or Complementary? Social Policy & Administration, 39 (7): 740-763.

Dahlström, E. (1969) Fördjupad företagsdemokrati. Prisma. Stockholm.

34

Dallner, M., Elo, A-L., Gamberale, F., Hottinen, V., Knardahl, S., Lind-ström, K., Skogstad, A., Örhede, E. (2000) Validation of the General Nordic Questionnaire (QPS Nordic) for Psychological and Social Factors at Work. Copenhagen. Nordic Council of Ministers.

Du Gay, P. & Salaman, G. (1992) The Cult[ure] of the Customer. Journal

of Management Studies, 29 (5): 615-633. Forssell, A. & Jansson, D. (2000) Idéer som fängslar – recept för en of-

fentlig reformation. Malmö. Liber Ekonomi. Gardell, B. (1981) Autonomy and participation at work. In Levi, L. (ed)

Society, stress and disease. Vol 4. London. Oxford University Press. Graefe, P. (2004) Personal services in the post-industrial economy: add-

ing nonprofits to the welfare mix. Social Policy & Administration, 38 (5): 456-469.

Gustafsson, R. (2000)Välfärdstjänstearbetet. Dragkampen mellan offent-

ligt och privat i ett historiesociologiskt perspektiv. Göteborg. Daida-los.

Gustafsson, R. & Szebehely, M. (2005) Arbetsvillkor och styrning i äld-reomsorgens hierarki – en enkätstudie bland personal och politiker. Rapport i socialt arbete, nr 114. Stockholm. Institutionen för socialt arbete.

Hoggett, P. (1996) New modes of control in the public service. Public

Administration, 74: 9-32. Hood, C. (1991) A public management for all seasons? Public Admini-

stration, 69: 3-19. Hood, C. (1995) The “New Public Management” in the 1980s: Variations

on a theme. Accounting, Organizations and Society, 20 (2/3): 93-109. Huzzard, T. (2003) The Convergence of the Quality of Working Life and

Competitiveness. A Current Swedish Literature Review. Stockholm. Arbetsliv i omvandling. Arbetslivsinstitutet.

Härenstam, A., Marklund, S., Berntsson, E., Bolin, M., Ylander, J. (2006)

Understanding the organisational impact on working conditions and health. Stockholm. Arbete och Hälsa. Arbetslivsinstitutet.

••

35

Jacobsen, D.I. & Thorsvik, J. (1998) Hur moderna organisationer funge-rar. Lund. Studentlitteratur.

Johnson, J. (1986) The Impact of Workplace Social Support, Job De-

mands and Work Control upon Cardiovascular Disease in Sweden. Baltimore, Department of Behavioural Science and Health Education, Johns Hopkins University and Stockholm, Department of Psychology, University of Stockholm.

Johnson, N. (1987) The welfare state in transition. The theory and prac-

tise of welfare pluralism. Brighton. Wheatsheaf. Kanter, R.M. (1993) Men and Women in the Corporation. 2nd ed. New

York. Basic Books. Karasek, R. (1979) Job Demands, Job Decision Latitude and Mental

Strain: Implications for Job Redesign. Administrative Science Quar-terly, 24 (2): 285-308.

Karasek, R. (1989) Control in the workplace and its health-related as-

pects. In Sauter, S.L., Hurrell, J.J. & Cooper, C.L. (eds) Job Control and Worker Health. Chichester. Wiley: 129-159.

Karasek, R. & Theorell, T. (1990) Healthy Work. New York. Basic Books. Küchen, T. (1994) Ny kooperation och den offentliga sektorn. In Aléx,

P., Normark, P., et.al (eds) Kooperation och välfärd. Stockholm. Fö-reningen Kooperativa Studier.

Larsson, K. (2006) Hemtjänst och anhörigvård. In Vogel, J. & Häll, L.

(eds) Äldres levnadsförhållanden. Arbete, ekonomi, hälsa och sociala nätverk 1980-2003. Stockholm. Statistiska centralbyrån / Umeå uni-versitet: 421-434.

Legge, K. (1995) Human Resource Management: Rhetorics and realities.

London. Macmillan. Lund, B. (1993) An Agenda for Welfare Pluralism in Housing. Social

Policy & Administration, 27 (4): 309-322.

36

Miettinen, A. (2000) Institutional characteristics of co-operatives provid-ing welfare services. Theoretical study of co-operatives and their members. Umeå. Umeå University Press.

Mintzberg, H. (1979) The Structuring of Organizations. Englewood

Cliffs, New Jersey. Prentice-Hall. Montin, S. (1997) New Public Management på svenska. Politica, 29:

261-278. Nilsson, J. (1986) Kooperativ utveckling. Lund. Studentlitteratur. Pestoff, V. (2000) Enriching Swedish Women’s Work Environment: The

Case of Social Enterprises in Day Care. Economic and Industrial De-mocracy, 21: 39-70.

Pugh, D.S. & Hickson, D.J. (1993) Great Writers on Organization. The

omnibus edition. Aldershot. Dartmouth Publishing Company Ltd. Rasmussen, B. (2004) Between Endless Needs and Limited Resources:

The Gendered Construction of a Greedy Organisation. Gender, Work and Organisation, 11 (5): 507-525.

Rothschild-Whitt, J. (1979) The Collectivist Organization: An Alternative

to Rational-Bureaucratic Models. American Sociological Review, 44: 509-527.

Runesson, I. & Eliasson-Lappalainen, R. (2000) Att sörja för äldre. En

översikt. Stockholm. Svenska Kommunförbundet, nr 4. Sjöstrand, S-E. (1993) Om företagsformer. Stockholm. Handelshögsko-

lan, EFI. SKTF (2002) 1000 chefer kan inte ha fel - om utvecklingssamtal, antal

medarbetare och andra förutsättningar för ett gott ledarskap. Stock-holm. Sveriges Kommunaltjänstemannaförbund.

Smith, V. (2006) Introduction: Worker Participation. Current Research

and Future Trends. In Smith, V. (ed) Worker Participation:Current Research and Future Trends. Research in the Sociology of Work, vol 16. Amsterdam. Elsevier.

••

37

SOU 1999:66 God vård på lika villkor. Slutbetänkande av kommittén om hälso- och sjukvårdsfinansiering och organisation. Stockholm. Fritzes.

Statskontoret, (1997) Stöd till kooperativ utveckling. Stockholm.

Statskontoret. Sturdy, A. & Korczynski, M. (2005) In the Name of the Customer: Ser-

vice Work and Participation. In Harley, B., Hyman, J., Thompson, P. (eds) Participation and Democracy at Work. Essays in Honour of Harvie Ramsay. New York. Palgrave Macmillan.

Szulkin, R. (1996) Organisationers struktur. In le Grand, C., Szulkin, R.

& Tåhlin, M. (eds) Sveriges Arbetsplatser – organisation, personalut-veckling, styrning. Stockholm. SNS Förlag.

Thorsrud, E. & Emery, F. (1969) Mot en ny bedriftsorganisasjon. Oslo.

Taunus. Trydegård, G.B. (2001) Välfärdstjänster till salu – privatisering och alter-

nativa driftsformer under 1990-talet. Välfärdstjänster i omvandling. SOU 2001:52.

Waldenström, K. & Härenstam, A. (2006) Hur skapas bra arbetsförhål-

landen? En studie av strategier hos chefer och anställda. Stockholm. Arbets- och miljömedicin.

Wallenberg, J. (1997) Kommunalt arbetsliv i omvandling. Styrning och

självständighet i postindustriell tjänsteproduktion. Stockholm. SNS Förlag.

Wallenberg, J. (2000) Konkurrens, bolagisering och ändrat huvudman-

naskap inom sjukvården. Konsekvenser för personalen. Vad säger forskningen? Stockholm. Stockholms Universitet, företagsekonomiska institutionen.

Wetterberg, G. (2000) Kommunerna. Stockholm. SNS Förlag. Wijkström, F. & Lundström, T. (2002) Den ideella sektorn: Organisatio-

nerna i det civila samhället. Stockholm. Sober Förlag.

•••